For this exercise you will explore couples' attitudes about work and family, how they handle household and childcare responsibilities, and the impact that these arrangements have on their relationship satisfaction. You will use crosstabulation and comparison of means.
Because the focus of this module is parents' attitudes and experiences, we recoded the variable "PERID_1" ("person type") to include only the categories "mom" and "dad." The new variable is called PARENT.
Attitudes about Work and Family
The variable PRVIDE_1 describes who respondents feel should provide the majority of the income in the family. We recoded it to exclude missing data and the category "other" and called the new variable PRVIDE.
Run a comparison of means of PRVIDE by PARENT. Remember that a code of 3 means the respondent feels that income should be provided equally. A code of less than 3 means the respondent feels his/her spouse should provide the majority of income, and a score of more than 3 means the respondent feels they should provide a majority of the family income.
What do the results say about mothers' and fathers' expectations for providing the family income?
The variable SPCHOR_1 measures attitudes about whether a husband should share equally in the household chores if his wife is working full time. To facilitate the analysis we recoded the variable to exclude missing data and collapsed the categories from five ("strongly disagree," "disagree," "neither agree nor disagree," "agree," "strongly agree") to three ("strongly disagree/disagree," "neither agree nor disagree," "agree/strongly agree"). The new variable is called SPCHOR.
Look at the crosstab of SPCHOR and PARENT. Are fathers more or less likely than mothers to agree that husbands should share equally in the housework? Do you find the results surprising?
Household Division of Labor-Self Reports
To see whether these attitudes are reflected in the actual household division of labor and work experiences, consider the variables which measure how much time, on average, each spouse spends doing a number of activities: shopping for the house (RSHPHS_1), taking kids to activities (RCHACT_1), cooking (RCOOKG_1), washing dishes (RDISHS_1), cleaning (RCLEAN_1), doing laundry (RLAUND_1), doing yard work and maintenance around the home (RYARDM_1), and helping kids with homework (RHPHWK_1). Each variable is coded into discrete categories ("1-2 hrs","3-5 hrs", etc.), which does not allow for the computation of means. To facilitate the analysis we recoded each of these variables so that respondents who answered "1-2 hrs" were coded as "1.5" hrs, those who answered "3-5 hrs" were coded as "4" hrs, and so on. We named the new variables "SHOP," "ACTIVITIES," COOK," "CLEAN," "DISHES," LAUNDRY," "HOMEFIX," and "HOMEWORK."
To further simplify the analysis, we created two indexes of time spent on activities in the home. HSWORK is the sum of SHOP, COOK, DISHES, CLEAN, and LAUNDRY, and measures how many hours per week respondents spend doing all of these chores. Similarly TIMEKIDS combines ACTIVITIES and HOMEWORK and measures how much time respondents spend driving their children to activities and helping them with homework each week. The variable HOMEFIX was left out of both indexes and will be analyzed on its own.
Run comparisons of means of HSWORK by PARENT, HOMEFIX and PARENT, and finally TIMEKIDS and PARENT. Note each analysis includes a filter to include only those respondents who work full time.
On average, how many hours do wives spend doing housework? How many hours do husbands spend on these chores? What do the results show about yard work and home maintenance? What about children-related activities? Where are gender differences the greatest? Where are they the smallest?
Household Division of Labor-Spouse Reports
Next consider men and women's assessments of their partners' time spent on household labor, measured by the variables SSHPHS_1, SCHACT_1, SCOOKG_1, SDISHS_1, SCLEAN_1, SLAUND_1, SYARDM_1, and SHPHWK_1.
Using the same coding scheme described in the previous analysis, we created new variables (SPSHOP, SPACT, SPCOOK, SPCLEAN, SPDISHES, SPLAUNDRY, SPHOMEFIX and SPHOMEWORK) and combined them into two indexes: SPHSWORK and SPTIMEKIDS.
Using a filter to limit the analysis to respondents who are employed full time, run comparisons of means of SPHSWORK by PARENT, SPHOMEFIX by PARENT, and SPTIMEKIDS by PARENT. When looking at the results, keep in mind that here the numbers in the "dad" cells refer to the number of hours men think their wives are spending on chores. Similarly, results in the "mom" cells refer to the number of hours wives believe their husbands are spending on chores.
According to husbands, how many hours on average do their wives spend doing housework, yard work, and childcare? According to wives, how many hours on average do their husbands spend on these chores? Are the results consistent with the findings from the analysis of respondents' self-reports? Where are the discrepancies the greatest?
Work and Childcare
Now think about gender differences in leaving the workforce to care for children. To simplify the analysis, we recoded the variable TMEOFF_1 into three categories: 1) never took time off; 2) took time off ; and 3) quit job . The new variable is called TIMEOFF.
Examine the crosstab of TIMEOFF and PARENT. What percentage of respondents report never having taken time off? What percentage of respondents say they had to quit their jobs to take care of their children? Which gender appears most likely to take time off?
Finally, examine whether respondents choose a job to be geographically closer to their children during the day. This is measured by the variable WKCHLD_1 which is coded into 4 categories ("not true at all," "somewhat true," "true," and "very true"). For this analysis, we collapsed the categories into two ("not true" and "true). The new variable is called WKCHLD.
Run a crosstab of WKCHLD and PARENT. What percentage of men and what percentage of women report choosing their job based on its geographical proximity to their children?
Relationship Satisfaction
Finally, consider the relationship between gender and relationship satisfaction. You will use two measures of relationship satisfaction: RESPAR_1, which asked respondents to agree or disagree with the statement, "I am happy with how we handle role responsibilities in our relationship;" and OVERAL_1, which refers to the statement, "Overall, I am satisfied with my relationship with my spouse/partner." We recoded both into new variables with three answer categories: 1) strongly disagree/disagree, 2) neither agree nor disagree, and 3) agree/strongly agree. The new variables are called RESPAR and OVERALL.
Run a crosstab of RESPAR and PARENT, using the filter RJBFT1_1(2) to limit the results to respondents who work full time. Which gender reports the highest level of satisfaction? What percentage of mothers is unhappy with the way role responsibilities are handled? Do you find the results surprising?
Finally, run a crosstab of OVERALL and PARENT (limited, once again, to respondents who work full time). Are respondents satisfied with their relationship with their spouse? Is one gender more satisfied/dissatisfied than the other? Do the results surprise you in light of the previous analyses?
